Autumn 2020 mainline Water-Trak demonstration trials
In October and November 2020, we completed a series of over-night trials of our Water-Trak system installed in a Northern Railway Class 319 train, operating from the Allerton, Liverpool depot. Figures 1 shows the train at St Helens station during one night of trials.
Fig 1: Northern Railway Class 319 train equipped with the Water-Trak system.
During four nights of trials on a section of mainline between Liverpool and Wigan (on the 28th of October and the 11th, 18th and 25th of November), the system was operational over 120 miles of signal protected track (in a Signal Protected Zone or SPZ). Multiple train passes were conducted over the same section of track to quantify the effect of water addition on following trains – see figure 2 for a schematic of the test route.
Fig 2: Water-Trak Autumn trial test route.
Before the trials, the preceding Rail Head Treatment Train (RHTT) pass was suspended to increase the chance of low adhesion conditions being encountered. The driver was instructed to follow the timings of a normal passenger service, stopping at every station along the route. Over the four nights, the adhesion risk level was rated between moderate and high. During the trials, the Water-Trak system was triggered both by operation of the Wheel Slide Protection (WSP) system and manually – the first ever demonstration of Water-Trak deployment on the mainline!
Train speed traces for the journeys between stations were analysed, looking at both acceleration and deceleration rates. As an example, figure 3 shows a speed trace for three laps of the test circuit between Garswood and Bryn on the 28th of October 2020
Fig 3: Train speed traces for three test laps between Garswood and Bryn
Table 1: 0-20 mph acceleration rates for three laps departing Garswood.
Looking at the departures from Garswood, the acceleration on lap 1 was poor and the train never reached full speed. On lap 2, the Water-Trak system was operated and the acceleration improved significantly. This improvement was maintained on lap 3 without the need for further water delivery. Table 1 shows the acceleration rates achieved for the three laps.
Table 2: deceleration rates for three laps approaching Bryn
As can be seen from the lap 1 speed trace, a large wheel slide occurred as the train braked approaching Bryn; this triggered WSP and caused the Water-Trak system to deploy. The train still achieved a relatively high deceleration despite the slide and the required braking distance was maintained. During braking on lap 2, the deceleration rate improved and the amount of slide reduced significantly. On lap 3, there was no slide at all and the level of deceleration was close to that encountered in good adhesion conditions. Table 2 shows the deceleration rates achieved for the three laps.
Across all the testing with Water-Trak operational, the lowest deceleration encountered was 7.4%g – note all the deceleration figures quoted here were achieved with step 3 braking. Much lower deceleration rates (<3%g) were encountered when water was not deployed. In all cases the Class 319 single fixed-rate sanding system was enabled.
The combined result of the improved braking and traction demonstrated in the trials is a reduction in journey time as shown in figure 3. An analysis was conducted on data from other Class 319 train journeys running over the same section of track. Figure 4 shows a comparison of journey times between Summer and Autumn.
Fig 4: Summer and Autumn journey times between Garswood and Bryn.
This analysis showed that in Autumn, the average journey time increased by 8.4 seconds and the variation in journey time doubled. The histogram of the Autumn data is right skewed with 6% of journeys incurring more that 30 seconds of delay. This analysis is for just one short journey between two stations; this effect accumulates over multiple stops, resulting in the significant rail industry disruption which is often encountered during Autumn.
To illustrate how the trial findings relate to the journey time data just described, they were overlaid onto the Autumn histogram – see figure 5.
Fig 5: Trial data overlaid on Autumn journey times.
The figure shows three examples from our testing. The good adhesion results occurred on a night with heavy rain – nature’s Water-Trak! The poor adhesion results were recorded on the “Down” line where Water-Trak was disabled – these occurred on an RHTT cleaned rail with dew forming (possible “wet rail” phenomenon). The Water-Trak results are from the speed/time traces in figure 3 and show the benefit of water addition on journey time. We would anticipate that, with further Water-Trak passes, the journey times would continue to converge on those recorded in good adhesion conditions.
In addition to the operational benefits described, the trials served to increase confidence in the system in a number of ways. Water-Trak was well received by drivers and inspired confidence in train braking; this is important because driver behaviour is a well-known contributory factor to Autumn delays. Between installation and removal of the Water-Trak systems, the train covered over 16,000 miles without any reported issues for train or infrastructure. Servicing the Water-Trak system in the depot proved to be straightforward, using existing water filling procedures and requiring less than 15 minutes to complete.
For more information about this project, please visit our First of a Kind 2020 demonstration video. We are now looking forwards to the next stage of our Water-Trak journey when we hope to deploy water addition from service trains carrying passengers on the mainline during Autumn 2021.